GPIC - GPU Power Iteration Cluster

TL;DR

GPIC is a GPU-accelerated clustering algorithm based on Power Iteration Clustering, significantly faster than existing Spark implementations while maintaining original properties.

Contribution

It adapts the Power Iteration Clustering algorithm to GPU architecture, enabling faster clustering without losing core features.

Findings

Achieved substantial speed-up over Spark implementations

Maintained the original properties of the PIC algorithm

Demonstrated effectiveness on test problems

Abstract

This work presents a new clustering algorithm, the GPIC, a Graphics Processing Unit (GPU) accelerated algorithm for Power Iteration Clustering (PIC). Our algorithm is based on the original PIC proposal, adapted to take advantage of the GPU architecture, maintining the algorith original properties. The proposed method was compared against the serial and parallel Spark implementation, achieving a considerable speed-up in the test problems.